# Failing Fast

Automatically cancelling your test suite right after the first failure is called a "fail-fast strategy". Currents dashboard allows tracking the outcome of your tests and automatically cancels runs whenever the first failed spec file is detected.

After executing a spec file, the runner sends the results to the Currents dashboard. The dashboard receives the results, identifies a failed test, and marks the associated run as "cancelled". Any new requests associated with the run will fail with a warning, as a result, cypress runners will not be able to start executing a new spec file for the run.

Please note, that any spec file that is still in progress will run to completion and its results will still be accepted.

![Example of a run automatically cancelled via "fail-fast" policy](https://3745692499-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FqmFDEiUa9mr11LUlxDnt%2Fuploads%2FbvW5UcQSyquIVkT8QrIA%2FScreenshot%202023-10-12%20at%2002.23.30.png?alt=media\&token=68cf4ac5-cffd-46ff-9cab-cc07ac319015)

Using a "fail-fast" strategy can be beneficial because:

* it reduces CI resources usage by preventing running additional tests after detecting just a single failure;
* potentially getting faster feedback for runs with a failed test - you don't need to wait for the run to finish;
* it reduces the number of tests results recorded on the Currents dashboard;
* it prevents [run-timeouts](https://docs.currents.dev/dashboard/runs/run-timeouts "mention") when an external tool (CI or a script) terminates cypress tests prematurely.

At the same time, utilizing a "fail-fast" strategy has some caveats:

* not all the tests will be executed - you will only see the results of a single test - the first test that triggered the "fail-fast" policy;
* tackling one failure at a time can result in more runs than necessary - one can invest more time and resources, compared to running all the tests regardless of their results.

![Automatically cancelled runs will be marked as failed and cancelled](https://3745692499-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FqmFDEiUa9mr11LUlxDnt%2Fuploads%2F3v1Yjpc2DSF3AlVCPRhk%2FScreenshot%202023-10-12%20at%2002.25.10.png?alt=media\&token=4c06da1d-225a-41a4-91f7-6952a28e2815)

If you are dealing with a stable and predictable suite of tests that rarely fail - "fail-fast" can be a great way to optimize your cypress tests.

{% hint style="info" %}
[Read more](https://docs.currents.dev/dashboard/runs/cancel-run) about how cancelling runs affects runs status, analytics and integrations
{% endhint %}

### Enabling fail-fast strategy

To enable the "fail-fast" strategy navigate to the **Manage Project** screen and toggle **Enable Fail Fast Strategy** controller.

Customers using our [cypress-cloud](https://docs.currents.dev/resources/reporters/cypress-cloud "mention") integration can control the "fail-fast" behaviour using the command line CLI flag `--auto-cancel-after-failures <number | false>`.

Customers using our [currents-playwright](https://docs.currents.dev/resources/reporters/currents-playwright "mention") integration can use `--pwc-cancel-after-failures <number | false>` option.

If set, it overrides the project's default fail-fast strategy setting. If not set, use the default project settings:

* `false` prevents fail-fast
* `number` aborts the run across all the participating machines failed + skipped tests detected exceed the provided value

For example

* `npx pwc --key xxx --project-id xxx --ci-build-id id-001 --pwc-cancel-after-failures 1`
* `npx cypress-cloud run --parallel --record --key xxx --ci-build-id id-001 --spec "./cypress/e2e/*.spec.js" --auto-cancel-after-failures`**`false`** will deactivate the fail-fast strategy, regardless of your project settings
* `npx cypress-cloud run --parallel --record --key xxx --ci-build-id id-001 --spec "./cypress/e2e/*.spec.js" --auto-cancel-after-failures`**`5`** will stop the run if more than 5 failed tests detected


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.currents.dev/guides/ci-optimization/fail-fast-strategy.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.